Previous efforts in chemoprevention the need for Phase II/III trials of intermediate size and duration which are designed to establish the efficacy of promising agents. Concomitantly, the need is to establish the validity of the most promising histopathologic and laboratory-based surrogate endpoint biomarkers SEB) currently held to be "reasonably certain" predictors of cancer prevention. One population to establish such goals consists of patients with oral leukoplakia. Leukoplakia is considered a pre-malignant lesion with a potential of developing malignancies approaching 30 percent depending on the degree of underlying dysplasia. Furthermore, these patients can be readily examined on a routine outpatient basis. Relevant to this population, a chemopreventive agent whose anti-cancer potential has been demonstrated by preclinical research efforts is curcumin, a naturally occurring derivative of the herb turmeric. Curcumin's anti-cancer effect is principally mediated through a topical effect and as an anti- inflammatory agent capable of modulating arachidonic acid metabolism. Therefore, in this multi-institutional two arm randomized placebo controlled trial, 20 patients will receive curcumin and placebo, respectively (total = 40 patients). Patients will be treated for a period of six months. Follow-up will be conducted for an additional six months following cessation of curcumin. Concomitantly, the relationship between curcumin-induced changes in leukoplakia and various SEB will be established. SEB will include measurement of mucosal proliferation index using computer-assisted image analysis of the S-phase fraction, as well as probes for proliferating cell nuclear antigen. Ploidy analysis will likewise be determined using computer-assisted image analysis. Concomitantly, the effect of curcumin on mucosal prostaglandin E2 will be established. The nuclear grading of leukoplakic tissue will likewise be assessed. Results of this trial will provide evidence of the value of curcumin as a chemopreventive agent and the significance of various SEB both as markers of disease risk as well as measures of biologic efficacy of curcumin as a chemopreventive agent.